Robert Shiller: The next bubble (or bust)

The stockmarket bust may not yet be over, says Robert Shiller, while an oil price boom might spark a slowdown – or even a crash – in soaring house prices –

Fluctuations in the world economy are largely driven by confidence. A changing level of public confidence is the ultimate driver behind much of the variation in individual and national incomes, in employment rates, in corporate earnings, in interest rates and in many other measures of the world economy.

Big changes in public confidence are often effected through speculative bubbles. It was a stockmarket bubble that kept the world economy humming through the late 1990s, and a housing bubble that kept the world economy from stumbling too badly after the stockmarket bubble burst in 2000. Where the world economy goes next will be largely determined by the nature and extent of the next bubble and how it bursts.

Bubbles have happened so often that we ought by now to have a science to understand them. We have the beginnings of such a science in the exciting emerging field of behavioural finance. Behavioural finance does not yet allow us to predict accurately where the next bubble will be, or when it will burst, but it helps us sort through the possibilities.

In a bubble, asset prices become the focus of attention, coordinating changes in confidence across millions of people. A bubble begins when initial price increases attract public notice. Economic theories that have lain dormant for years are suddenly trotted out to justify the bubble, gaining new prominence when reported in the media. Given the “oxygen of publicity”, the initial price rises generate investor excitement and, supported by the new theories, feed back into further price increases, generating yet more excitement, in what appears to be a virtuous cycle.

Bubbles: a history

Like real bubbles, asset-price bubbles are insubstantial and prone to burst. This is because high prices are supported largely by expectations of further, rapid, price increases. But rapid price increases cannot go on for ever, because they require repeated increases in demand to be sustained. A bubble will collapse of its own internal dynamics eventually, without any exogenous bad news. The apparently virtuous cycle of rising prices is then replaced by a downward spiral of prices and confidence.

The graph shows the history of world prices in the equity, housing, and oil markets. Are the big fluctuations seen here speculative bubbles? I believe they are. None of these extreme movements in stock prices, for example, has ever been justified by economic fundamentals.

There was a dramatic world stockmarket boom from 1982 to 2000, with a six-fold increase in price in real (that is, after-inflation) terms. This occurred in two episodes: the first, 1982 to 1989, dominated by Japan, and then from 1995 to 2000, dominated by America and Europe. Many economic theories that might justify these price increases were propelled by the latter boom. The most prominent was that new technology, notably the Internet, had created a “new economy” of permanent higher productivity growth. But from March 2000, the world stock market fell by a half to its nadir in 2002, before rebounding somewhat. The end of the bubble in 2000 was not caused by any change in the technology supposedly underpinning this “new economy”, but simply by the internal dynamics of the bubble.

Safe as houses?

House prices in many glamour cities took off around 1997. An index of prices in London, Paris, Moscow, Shanghai, Sydney and Los Angeles is shown in the graph, beginning in 1983 (data before that date are sketchy). These cities boomed in the late 1980s, crashed in the early 1990s and then boomed again. Real house prices are now almost twice what they were in 1997. There is talk that the rate of increase in some cities, such as London, is slowing, but no clear signs yet that the bubble is bursting.

What is behind this house price boom? It has been driven in part by the fall in stockmarkets since 2000. Financial scandals like Enron and WorldCom also dented confidence in equities. Terrorist attacks, especially those of September 2001, also seem to have left investors with a desire to invest in something tangible. A more nebulous story favoured in boom cities themselves is that they have a unique intellectual, commercial and technological mix that makes them winners in the new global economy. Celebrity culture – fostered by the information society – also seems to play a part in driving fashionability and, therefore, prices in cities from Dublin to Shanghai.

Oil’s not well

One might suspect that oil price volatility resulted from exogenous supply disruptions and inelastic demand. But it is not so simple. True, oil prices rose abruptly between 1973 and 1974 (the first oil crisis, thanks to the supply squeeze by OPEC, the oil-producers’ cartel) and between 1979 and 1980 (the second oil crisis, due to the Iranian revolution and the Iran-Iraq War). But oil remained expensive from 1974 to 1986.

So what else was going on – besides politics and war – to cause oil prices to remain high during this period? For starters, there was the “great population scare”. The 1972 report Limits to Growth by the Club of Rome, a think-tank, sparked fears of a looming economic catastrophe as natural resources were used up. The scare affected both the oil and equity markets.

Starting in 1998, with a brief interruption in 2002, oil prices increased four-fold. Many attributed this to demand from fast-growing developing markets like China and India. Moreover, there are fears that, as this growth continues, they will make even more extraordinary demands on natural resources like oil. These fears have pushed prices still higher.

Related rises?

Do any of these booms cause booms – or indeed, busts – in other assets? As the graph shows, the three series are for the most part independent of each other, though oil and equities seem to move in opposite directions. The sudden doubling in real oil prices thanks to a surprisingly strong increase in oil demand in the late 1940s roughly coincides with a halving of real stock prices. The high oil prices of 1974 to 1986 match a period when the stockmarket was distinctly below trend. After 1986, world stock prices boomed as oil prices were low and generally falling. After the turn of the millennium, the oil market has been generally rising, and the stockmarket generally falling.

One might expect a relation between oil and house prices, since high oil prices tend to damage the economy and hence people’s ability to pay for their homes. They also raise the costs of heating a home. However, over the observable time period, there is no obvious relation. There ought to be at least some relation, though, since an oil price boom can create a recession, and recessions tend to be bad for housing markets.

The next bubble or bust

The stockmarket is the hardest to predict. It is not safe to say that the sharp falls since 2000 are over. Some markets are still high. Nor can we say, as some do, that the experience of a big market fall has rid us of our bubble mentality, making another such boom impossible.

We could have both a new equities boom and then another severe bust in the near future. The 1920s’ stockmarket expansion peaked in 1929 and crashed sharply: the market had lost 63% of its 1929 peak real value by 1932, a bigger drop than between 2000 and 2002. But, contrary to popular recollection, the 1929 crash did not curb speculative impulses. There was another boom, starting in 1932, spurred by apparent evidence of a recovery from the Great Depression. This boom, also visible in the figure, peaked in 1937. This now-forgotten boom was just as big as the boom of the 1920s. And after it the stock market fell as far as it did after 1929, though it took much longer – until 1949.

A house of cards?

The housing market is more predictable than the stockmarket. Unlike stock prices, which change randomly, house prices display positive serial correlation: they tend to move up and up and up till they turn, and then they move down, down, down. Even more than stock prices, housing markets are governed by “momentum” and sentiment. Therefore, given recent rises, one might expect further increases. But the real estate boom is showing signs of fizzling out in some cities. We have learned from experience that a slowdown often precedes a price collapse. Moreover, despite the lack of evidence of any link between them, the oil price boom could threaten the world economy, thus bringing with it the end of the housing boom.

Oil price rises are generating a resurgence of public fears about the oil market, especially the voracious energy appetite of India and China. The sudden public recognition that there could be binding resource constraints as emerging countries develop could encourage potential oil suppliers to hold off on development so that they can sell at higher prices later. But in itself this would create higher prices today, sparking a prolonged speculative bubble like that from 1974 to 1986. Such a bubble could spell real trouble for the stock and housing markets.

Indeed, even though oil prices have fallen from their peak in October, the risk of such a bubble in oil and other commodities in coming years, which could set off retrenchment, might be the biggest threat to the world economy. There is a danger that, instead of looking at ways of reducing consumption or for alternative sources, we will become transfixed by the risks of high prices to a rapidly growing world economy.

Ironically, this could inhibit that very growth for years to come.

Robert Shiller
Robert J Shiller is Stanley B Resor Professor of Economics, Cowles Foundation and International Center for Finance, Yale University, and co-founder of Macro Securities Research, LLC. He is author of Irrational Exuberance, Princeton, 2000 and New Financial Order, Princeton, 2003.